Cyber Resilience

CVE-2026-54233

Vllm ≤ 0.23.1

Published
22 June 2026
Modified
24 June 2026
Patch / advisory
CVSS Score v3.1 6.5
Click a component to see what it means
Raw vectorCVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
EPSS Score 0.0042 35th percentile
Risk Priority 50 floored blend · peak EPSS

Summary

CVE-2026-54233 is a medium-severity Data Amplification (CWE-409) vulnerability in Vllm Vllm. Its CVSS base score is 6.5 (Medium).

Operationally, exploitation aligns with the MITRE ATT&CK technique Endpoint Denial of Service (T1499); ranked at the 35th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog.

The strongest mitigations our analysis identified map to SI-10 (Information Input Validation) and SC-5 (Denial-of-service Protection) — see the control section below for these in your framework.

EU & UK References

Vulnerability Data

vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.23.1rc0, vLLM's /v1/audio/transcriptions endpoint limits compressed upload size but not decoded PCM output. A 25MB OPUS file expands to ~14.9GB of float32 PCM at decode time.…

more

This vulnerability is fixed in 0.23.1rc0.

CWE(s)

Related Threats

MITRE ATT&CK Enterprise Techniques

T1499 Endpoint Denial of Service Impact
Adversaries may perform Endpoint Denial of Service (DoS) attacks to degrade or block the availability of services to users.
T1499.003 Application Exhaustion Flood Impact
Adversaries may target resource intensive features of applications to cause a denial of service (DoS), denying availability to those applications.
Derived from this CVE’s CWE(s) via the direct CWE→ATT&CK cross-walk.

CVEs Like This One

CVE-2026-55574Same product: Vllm Vllm
CVE-2025-71379Same product: Vllm Vllm
CVE-2025-46560Same product: Vllm Vllm
CVE-2025-48887Same product: Vllm Vllm
CVE-2026-55646Same product: Vllm Vllm
CVE-2026-54234Same product: Vllm Vllm
CVE-2026-53923Same product: Vllm Vllm
CVE-2026-47155Same product: Vllm Vllm
CVE-2026-54236Same product: Vllm Vllm
CVE-2026-54235Same product: Vllm Vllm

Affected Assets

vllm
vllm
≤ 0.23.1

Mitigating Controls

Mitigating Controls (NIST 800-53 r5) AI

Input validation can reject or limit decompression of data whose expansion ratio exceeds safe thresholds.

DoS protection limits the resource-exhaustion impact when a decompression bomb is processed.

Resource allocation controls bound memory/CPU consumption that a data-amplification attack would otherwise exhaust.

Mitigating Controls (NIST CSF 2.0) AI

Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.

PR.PS-06 mostly match
prevents

Secure-development practices include input-validation and resource-limit checks that prevent improper handling of compressed data.

DE.CM-09 partial match
prevents

Runtime monitoring of compute resources can detect exhaustion caused by decompression bombs.

PR.IR-04 partial match
prevents

Capacity planning and monitoring directly limits the availability impact of data-amplification attacks.

Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI

Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.

finds

Security testing can uncover decompression-bomb vulnerabilities before release.

mitigates

Redundancy helps availability but does not address the root cause of the weakness.

finds

Monitoring can detect anomalous resource usage but does not prevent the weakness.

prevents

Secure development lifecycle includes input validation and resource-limit checks that mitigate data-amplification attacks.

prevents

Application security requirements can mandate limits on decompression size and ratio.

prevents

Secure architecture principles encourage defensive design against resource-exhaustion threats.

References